Xueshuantong Injection in Treating Deep Venous Thrombosis: A Systematic Review and Trial Sequential Analysis

Aims In recent years, the incidence of deep venous thrombosis (DVT) presents an increasing trend year by year. The current evidence regarding the efficacy and safety of Xueshuantong injection for DVT is controversial. This systematic review (SR) aimed to assess the efficacy and safety of Xueshuantong injection in the treatment of DVT systematically and provide an evidence-based reference for clinical treatment. Methods Nine electronic databases were used to identify the literature consisting of randomized controlled trials (RCTs) with a date of search of 1 November 2020. Clinical effective rate and incidence rate of adverse events were investigated as primary outcomes. Patency rate of femoral vein, patency rate of popliteal vein, patency rate of posterior tibial vein, circumference difference, activated partial thromboplastin time (APTT), and D-dimer (D-D) were investigated as secondary outcomes. Revman 5.4.1 was used to analyze the results. Analysis of the power of evidence was performed with Trial Sequential Analysis (TSA). Results A total of 12 articles including 1018 patients were included. The results of the meta-analysis showed that the clinical effective rate in the experimental group was higher than that in the control group, the incidence rate of adverse events in the experimental group was higher than that in the control group; after the operation, the patency rate of femoral vein, patency rate of popliteal vein, patency rate of posterior tibial vein, circumference difference, APTT, and D-D in the experimental group were significantly improved compared with those in the control group, and the difference between the groups was statistically significant. TSA suggested that the meta-analysis concerning the clinical effectiveness of Xueshuantong injection in the treatment of DVT was of adequate power to reach firm conclusions. Conclusion Based on the current analysis, Xueshuantong injection as an add-on treatment provided better treatment effect for DVT with adequate power but this benefit should be considered with caution because of the small number of studies included in the meta-analysis and the high or unclear risk of bias of the included trials, suggesting that further studies are needed.


Introduction
Deep venous thrombosis (DVT) is a common critical disease with very high incidence in elderly inpatients. Studies have shown that almost 20% of patients scheduled for orthopedic surgery have DVT, despite the use of preventative strategies [1,2]. In addition, if patients with DVT do not receive timely and effective treatment, almost 50% will develop DVT syndrome, significantly affecting their work and life, and 6% will die of pulmonary embolism [3,4]. Some scholars of traditional Chinese medicine consider the pathogenesis of DVT to be related to disorders of Qi and blood, and that the first line of treatment involves the promotion of blood flow. Xueshuantong injection is a safe and effective traditional Chinese medicine which is widely used in the treatment of various thrombus-related diseases including DVT [5]. While clinical studies in recent years have shown that Xueshuantong injection can treat DVT effectively, reducing symptoms with few adverse events, its effectiveness and safety have not been comprehensively and systematically evaluated.
is study systematically evaluated the efficacy and safety of Xueshuantong injection in treating DVT, in order to provide an evidence-based medical basis for the treatment of DVT.

Materials and Methods
e protocol for this systematic review has been developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [6].
It has been also registered on the INPLASY database (registration number: INPLASY2020120117).

Search Strategy.
e Chinese databases include Chinese National Knowledge Infrastructure Database (CNKI); the Chongqing VIP Database; Chinese Biomedical medical Database (CBM); and Wanfang Database.
e English databases include PubMed, Cochrane Library, Web of Science, Embase, and MEDLINE. e search terms are "Panax notoginseng saponins"; "Panaxnotoginseng extract"; "Xueshuantong"; "XUESAITONG injection"; "Deep Venous rombosis"; "DVT"; "efficacy"; and "safety". e search time is from the establishment time of each database to November 1, 2020. e complete search strategy for PubMed is reported in Table 1. e search strategy for PubMed will be adapted for use in other databases.
Add search is the search by mainstream search engines (Baidu, Google); Clinical Research Registry (WHO International Clinical trial Registration platform; China Clinical trial Registration Center).
At the same time, the two researchers searched the literature independently, cross-checked the results, and further searched the references according to the search results.

Research Type.
e type is randomized controlled trials (RCTs).

Study Population.
e original study clearly pointed out that the included patients were in line with the relevant diagnosis of DVT. Refer to the guidelines for diagnosis and treatment of deep vein thrombosis [7] (according to the symptoms, clinical signs, blood coagulation tests, color Doppler ultrasound results, etc).

Intervention.
e experimental group received Xueshuantong injection or Xueshuantong combined with conventional treatment (anticoagulation, thrombolysis, pressure treatment, and other clinical guidelines recommended treatments).

Outcomes
(1) Primary Outcomes. Clinical effective rate (the circumference difference of the lower limbs was less than 2.5 cm). Refer to the standard for diagnosis and efficacy of TCM Internal Medicine [8].
Incidence rate of adverse events: (2) Secondary Outcomes

Data Extraction.
e two researchers independently screened the included literature and extracted the data. e differences encountered in the process can be resolved through negotiation, or they can consult experts in the research team to record the process. Evaluation items are as follows: how random sequences are generated; whether distribution is hidden; whether researchers and subjects are blinded; whether study outcomes are blinded; whether outcome data remain intact; selective reporting; and whether there are other biases.

Statistical
Analysis. RevMan 5.4.1 was used for data analysis and odds ratio (OR) or relative risk (RR) was used for the dichotomous variable and mean difference (MD) or standardized mean difference (SMD) was used for continuous variables and 95% CI was used to test the effect. e heterogeneity was analyzed by using the Q value test and I 2 test. If P > 0.1 and I 2 ≤50%, no heterogeneity was judged, and the fixed effect model was used for metaanalysis. P > 0.1, I 2 >50% indicate heterogeneity; random effect model is used for meta-analysis; and the source of heterogeneity is found through sensitivity analysis or subgroup analysis (a sufficient number of studies are included). If the heterogeneity cannot be solved, the meta-analysis will be abandoned and descriptive analysis will be used.

Results of the Search.
A total of 330 potentially corresponding studies were identified by our primary search. After removing duplicate records, we identified 25 studies from 330 studies for review of the full text. A total of 13 studies were exempted for the following reasons: 5 articles are not RCTs, 4 articles are inappropriate types of intervention, 4 articles are repetitive publication, 1 article has not full text and 1 article is not peer-review paper. Of the 12 included studies, 1018 patients were used in this metaanalysis, as is shown in Figure 1.

Basic Characteristics of Included Studies.
e basic characteristics of the 12 RCTs are summarized in Table 2.

Assessment of Study Quality and Risk of Bias.
Of the 12 studies, 8 [9-12, 14-16, 20] reported random method, 6 studies [9-11, 14, 16, 20] used the random number table to generate random numbers; 2 studies [12,15] generated random numbers with a high risk of bias; and none of the studies reported allocation concealment. Figures 2 and 3 show the risk of bias summary of the included studies.

Clinical Effective Rate.
Eight studies [10][11][12][15][16][17][18][19] reported clinical effective rate, with a total of 739 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of clinical effective rate in the treatment of DVT, and the difference was statistically significant (OR � 4.68, 95% CI [2.79, 7.86], P < 0.00001). Considerably, there was no heterogeneity among the studies (P � 0.92, I 2 � 0%), and a fixed effect model was used for meta-analysis, the results of which are shown in Figure 4.

Incidence Rate of Adverse Events.
Six studies [9,10,16,17,19,20] reported the incidence rate of adverse reactions, with a total of 541 patients. No serious adverse reactions occurred in all the studies. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of the incidence rate of adverse events in the treatment of DVT, and the difference was statistically significant (OR � 0.48, 95% CI [0.27, 0.84], P � 0.01).
Considerably, there was no heterogeneity among the studies (P � 0.49, I 2 � 0%), and a fixed effect model was used for metaanalysis, the results of which are shown in Figure 5.

Patency Rate of Femoral Vein.
ree studies [11,13,20] reported patency rates of femoral vein, with a total of 212 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of patency rates of the femoral vein in the treatment of DVT, and the difference was statistically significant (OR � 1.45, 95% CI [0.80, 2.63], P � 0.23). Considerably, there was no heterogeneity among the studies (P � 0.51, I 2 � 0%), and a fixed effect model was used for meta-analysis, the results of which are shown in Figure 6.

Patency Rate of Popliteal Vein.
ree studies [11,13,20] reported patency rates of popliteal veins, with a total of 212 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of patency rates of popliteal veins in the treatment of DVT, and the difference was statistically significant (OR � 1.59, 95% CI [0.81, 3.13], P � 0.18). Considerably, there was no heterogeneity among the studies (P � 0.54, I 2 � 0%), and a fixed effect model was used for meta-analysis, the results of which are shown in Figure 7.

Patency Rate of Posterior Tibial Vein.
ree studies [11,13,20] reported patency rate of the posterior tibial vein, with a total of 212 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of patency rate of the posterior tibial vein in the treatment of DVT, and the difference was statistically significant (OR � 1.08, 95% CI [0.93, 1.25], P � 0.33). Considerably, there was no heterogeneity among the studies Evidence-Based Complementary and Alternative Medicine (P � 0.39, I 2 � 0%), and a fixed effect model was used for meta-analysis, the results of which are shown in Figure 8.

Circumference Difference of Upper Knee Limb.
Four studies [12,14,15,19] reported a circumference difference of upper knee limb, with a total of 312 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of circumference difference of upper knee limb in the treatment of DVT, and the difference was statistically significant (OR � −1.25, 95% CI [−1.37, −1.12], P < 0.00001). Considerably, there was no heterogeneity among the studies (P � 0.96, I 2 � 0%), and a fixed effect model was used for meta-analysis, the results of which are shown in Figure 9.

Circumference Difference of Lower Knee Limb.
Four studies [12,14,15,19] reported circumference difference of lower knee limb, with a total of 312 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of Circumference difference of lower knee limb in the treatment of DVT, and the difference was statistically significant (OR � −1.08, 95% CI [−1.18, −0.98], P < 0.00001). Considerably, there was no heterogeneity among the studies (P � 0.43, I 2 � 0%), and a fixed effect model was used for meta-analysis, the results of which are shown in Figure 10.

Activated Partial
romboplastin Time. Six studies [9,10,[15][16][17]19] reported APTT, with a total of 553 patients. e results showed that Xueshuantong injection was superior to conventional treatment in the improvement of APTT in the treatment of DVT, and the difference was statistically significant (OR � 3.52, 95% CI [1.73, 5.32], P � 0.00001). Considerable heterogeneity was found between the studies (P < 0.00001, I 2 � 91%), and a random effects model was used for meta-analysis, the results of which are shown in Figure 11.
After excluding the two studies [10,15], the heterogeneity of APTT was significantly reduced (P � 0.97, I 2 � 0%), and the two studies were judged as the source of heterogeneity, the results of which are shown in Figure 12 No full text (n = 1) Non peer-review paper (n = 1) Studies included in qualitative synthesis (n = 12) Studies included in quantitative synthesis (meta-analysis) (n = 12)                   Figure 13.

Publication Bias.
A maximum of 8 outcome indices were available so no assessment of publication bias (e.g., funnel plot) could be made and the possibility of this form of bias cannot be excluded.
3.6. Sensitivity Analysis. Sensitivity analysis was carried out by changing the statistical model and removing the recombination effect of a large sample study. is process did not change the results, indicating that the results of this study were robust.

Trial Sequential Analysis.
e type I and type II errors were set at 0.05 and 0.2, respectively. e sample size was calculated with statistical power at 80%. e clinical effectiveness of Xueshuantong injection in the treatment of DVT was analyzed by experimental sequence analysis. e value (curve) crossed the traditional boundary value after the second study and crossed the TSA boundary value after the fourth study, indicating that the positive conclusion had been obtained before the amount of sample information in the current cumulative study had not reached the expected   Evidence-Based Complementary and Alternative Medicine amount of information; that is, Xueshuantong injection is more effective than conventional treatment in the treatment of DVT, and that no further studies were needed for verification, as shown in Figure 14.

Discussion
Deep venous thrombosis refers to abnormal blood flow and coagulation in the deep venous system [21]. Clinical patients have lower limb swelling, local skin temperature rise, and pain, and may suffer fatal pulmonary embolism in severe cases [21][22][23][24]. In traditional Chinese medicine, this condition is referred to as "thigh swelling" and treatment is based on promoting blood circulation, removing blood stasis, and replenishing qi. Xueshuantong injection is a traditional Chinese medicine injection made from total saponins extracted from Panax notoginseng, a modern form of Chinese patent medicine [25,26]. e main components of Xueshuantong injection are ginseng and Panax notoginseng. Panax notoginseng has the effect of dispersing blood stasis and stopping bleeding and pain and detumescence. Panax notoginseng can reduce the levels of CD62p, CD63, GPIIb/ IIIa, FIB, and D-dimer in plasma of patients with deep venous thrombosis and thus play a therapeutic role. Ginseng preparation can also improve blood hypercoagulable state [27]. In addition, some studies have shown that Xueshuantong injection may have an antithrombotic effect by inhibiting platelet aggregation, improving blood flow, and maintaining an environment conducive to normal blood flow [28][29][30].
e results of this study show that Xueshuantong injection can treat deep venous thrombosis effectively, with fewer adverse reactions and significantly better outcomes (including deep vein patency, circumference of lower extremities, D-dimer, and APTT) than conventional treatment. e results of the sequential analysis showed that the results of this meta-analysis are robust and may therefore have implications for evidence-based clinical practice. However, in view of the small sample size included and the low quality of literature methodology, the strength of this finding is limited, and more high-quality studies are needed to improve the level of evidence in the future.
Limitations of this study: this study compares Xueshuantong injection with conventional treatment, the latter encompassing anticoagulation, thrombolysis, and physiotherapy, which may have introduced heterogeneity. In addition, the included studies were all sourced from Chinese research databases. No relevant research was found in English language databases, and the representation of only one sector of the global population may have introduced bias. In the process of literature screening in this study, several articles were excluded because key data could not be obtained by the author, which may have biased the results. In addition, the quality of the research included in this study was low, with small sample sizes in some cases, which further reduces the certainty of conclusions.
Future prospects: future clinical research should use homogeneous outcome indicators to gradually establish a core outcome index set with indicators related directly to patient benefits. Higher quality research is needed with high integrity, standardized reporting, and reduced bias.

Conclusion
Based on the current analysis, compared with control, Xueshuantong injection as an add-on treatment provided better clinical efficiency for DVT with adequate power, while with fewer adverse reactions and significantly better outcomes (including patency of deep vein, circumference of lower extremities, D-D, and APTT), but this benefit should be considered with caution because of the small number of studies included in the meta-analysis and the high risk of bias of the included trials, suggesting that further studies are needed.

RCTs:
Randomized clinical trials CNKI: China national knowledge infrastructure VIP: Chinese scientific journal database CI: Confidence interval MD: Mean difference RR: Relative risk DVT: Deep venous thrombosis PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses SMD: Standardized mean difference APTT: Activated partial thromboplastin time D-D: e D-dimer OR: Odds ratio TSA: Trial Sequential Analysis RIS: Required information size.

Data Availability
e data used to support the findings of this study are included within the article. Disclosure e funders had not been involved in the design, execution, or writing the study.

Conflicts of Interest
e authors declare no conflicts of interest.

Authors' Contributions
Wenhui Li, Changgeng Fu and Feng Xu contributed equally to this work and are co-first authors. WJF and CGF conceived the article, drafted the research protocol, retrieved the literature, analyzed the data, and wrote this manuscript. WHL and FX screened studies and evaluated the risk of bias. LX and XHW extracted data and gave suggestions for the discussion. RYH gave suggestions on the structure of the article. FX provided methodological guidance and gave suggestions on the conception of the article. All authors have read and approved this manuscript. 10 Evidence-Based Complementary and Alternative Medicine